Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
fooker
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
33 ms
·
721.
▲
by
fooker
10mo ago
> it isn't credible that B-2s flew over Fordow without Iranian permission It is very credible because a few days before this happened, Israel wiped out Iran's air defense systems with f35 attacks.
722.
▲
by
fooker
10mo ago
Just because you don't see these in the US doesn't mean they aren't elsewhere in the first world countries.
723.
▲
by
fooker
10mo ago
Being early is often the same as being wrong. LLMs could be a dead end, but aren't anywhere close to saturating the technology yet.
724.
▲
by
fooker
10mo ago
List membership check being significantly slower than set membership check is freshman computer science 101.
725.
▲
by
fooker
10mo ago
Throw in a couple of barriers.
726.
▲
by
fooker
10mo ago
Yes, maybe. Modern games already employ a bunch of VM-like techniques for tamper protection. This has effectively killed PC game piracy.
727.
▲
by
fooker
10mo ago
You can switch out the kernel in the running Linux desktop.
728.
▲
by
fooker
10mo ago
The last remaining roadblock is kernel level anti-cheat frameworks. Pretty horrible technology, and unfortunately a good majority of the gaming industry by revenue relies on it.
729.
▲
by
fooker
10mo ago
Yes, that's what I said. If you are writing performance sensitive code that is not covered by a popular Python library, don't do it unless you are a megacorp that can put a team to write and maintain a library.
730.
▲
by
fooker
10mo ago
> you need to know these numbers on Python to know when to actually build something in C People usually approach this the other way, use something like pandas or numpy from the beginning if it solves your problem. Do not write matrix mul
731.
▲
by
fooker
10mo ago
Counterintuitively: program in python only if you can get away without knowing these numbers. When this starts to matter, python stops being the right tool for the job.
732.
▲
by
fooker
10mo ago
Okay, get outraged. Not sure about the strawman ballroom argument but: Let me repeat, shouting "stupid Trump is horrible for the economy" from rooftops is not a great political strategy. In the case it turns out to be false, (and
733.
▲
by
fooker
10mo ago
I get that impression too. The way our world works, that would be completely fine for the 'economy looks set to accelerate' like in this article. A good economy is not necessarily good for the general public.
734.
▲
by
fooker
10mo ago
> Doesn’t sound like an obvious winning recipe. I agree. However, hammering the message of "this would obviously never work" is harmful. Because if it does work, half the country is never going to listen to you again, and the s
735.
▲
by
fooker
10mo ago
Some people genuinely believe Portland is a warzone. Your idea of 'rural areas' is not far off from that.
736.
▲
by
fooker
10mo ago
These things are not incompatible. My point is: assume evil, not stupid. Even when it's really tempting to think your political opponents are stupid.
737.
▲
by
fooker
10mo ago
Man it's not that deep. More factories -> economic boost. Whether it'll work is a different topic, but it seems silly to wish failure.
738.
▲
by
fooker
10mo ago
Several actually. https://www.finance.senate.gov/chairmans-news/the-one-big-be... https://www.congress.gov/bill/119th-congress/house-bill/2480 https://www.congress.gov/bi
739.
▲
by
fooker
10mo ago
California is not a representative slice of America. You could imagine turning a few smaller states’ economies to be more like China (from the current government subsidized agriculture business) without messing with the tech and finance ind
740.
▲
by
fooker
10mo ago
What if it does work though? It’s really dangerous to assume your political enemies are stupid and incompetent, even when they likely are. Because if it turns out you are wrong, you can only cry wolf a few times at best. The current adminis
741.
▲
by
fooker
10mo ago
Weird hill to die on in 2025 If you had said this in 2015, we would be nodding along
742.
▲
by
fooker
10mo ago
> private investment in business ventures with risk long predates limited liability being attached to the corporate form Here you go - https://en.wikipedia.org/wiki/Bottomry
743.
▲
by
fooker
10mo ago
> Limited liability is the government interfering I agree. That was not the point I was making. The government interferes here for good reason, as no one would start a business if they would be personally responsible for random things. T
744.
▲
by
fooker
10mo ago
> These companies make good profits They don’t make good profits. TSMC has fairly mediocre numbers by the standard of the tech industry. Intel has really bad numbers for the last several decades. AMD was having so much trouble with found
745.
▲
by
fooker
10mo ago
The alternative has been the government controlling which ventures are funded, at which point the government assumes the risk. Would you rather the only entity with a monopoly on violence be liable for all risky ventures? What happens whe
746.
▲
by
fooker
10mo ago
I did not say Nvidia does not have risk. I specifically meant this circular investment idea is medium risk. Of course Nvidia can tank if sales decline.
747.
▲
by
fooker
10mo ago
It’s a medium risk high reward strategy. You only fail here if all the customers fail, which admittedly there’s a chance of. If one succeeds, and grows 10-100x, you are profitable. If a few succeed, you double your market cap.
748.
▲
by
fooker
10mo ago
Money flowing in a circular manner is the definition of economy. This is the reason we moved to using money instead of a barter system.
749.
▲
by
fooker
10mo ago
Realistically, identify the next technology that will need such wizards and get into a PhD program to research that technology.
750.
▲
by
fooker
10mo ago
It’s harmful until you are the one getting personally sued by random grifters. We limit liability for risky ventures for a reason.
More ›